Institutional Repository of Vilnius University
Skraidančių paukščių aptikimas ir išbaidymas /
Abstract
dc:descriptionThe ability to effectively track and scare birds in real time using computers can be applied in many areas, especially if the tracking can be applied from great distances. However, bird-tracking in computer vision is a challenging task. Due to the size, movement characteristics, shape, color, and background of birds, tracking birds is a more complex task than, for example, human tracking. The main goal of this work was to create a bird tracking model and a robot controlled by it. In this work, we investigated the challenges of bird tracking and tried some popular object detectors using neural networks. Also, various image processing and tracking methods in the OpenCV library were tried. To create the bird-tracking model OpenCV image processing library and YOLO object detector was used. Then an Arduino robot was modeled and assembled to test the bird-tracking model, and a way to communicate between the robot and the model was created. The results of the bird-tracking model and robot have shown that real-time tracking of small birds is a complex task but possible even with average computer equipment.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Pazdrazdis, Laurynas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146237041&prefLang=en_US